cowteriyaki

I find the fact that beer pipelines have their own color designation in the map legend intriguing. Are they common enough outside of breweries to merit singling them out?

Coming with a Lidar out of the box seems nice.

Does the MARS hardware really remove the hidden extras (computer with a gpu) mentioned as the downside of HF SO-101 or LeKiwi? While a jetson is good for inference, I feel like to train VLAs you would need access to a powerful GPU regardless. For Lerobot based hardware training ACT was relatively low profile if you use low resolution for the camera feeds, but with increased resolution or with more than one camera I already saw needing more than 8GB of VRAM. If VLA is on the table, finetuning something like the open sourced version of pi0 should already necessitate access to more than one 4090 or above I think.

Also, do you have plans for community-level datasets? I think Lerobot sort of does this with their data recording pipeline and HF integration.

You are thinking of Chinese characters. Unlike the Japanese where knowledge of Kanji is pretty much required to read texts above a certain level, they are optional in Korean.

It is true that a good portion of vocabulary are derived from Chinese words (hence they /can/ be spelled with Chinese characters), and maybe 30+ years ago it was common practice to have newspapers and textbooks writing complex words in Chinese characters. This is no longer the case, and pretty much all texts these days spell everything with the Korean alphabet. The practice of learning non-simplified Chinese characters of course is still common among Koreans for several reasons.

I'm intrigued by the electrodes they claim to be having 4 times better SNR than conventional electrodes. AFAIK there has been numerous attempts at dry electrodes (electrodes w/o gel for impedance control), but due to higher impedance, SNR has always been worse than wet electrodes as a result. The main advantage of dry electrodes has been faster prep time & convenience at a cost of signal fidelity. I'd be interested in technical details, but information on their website doesn't really let on much. Wonder how they actually compare to market leaders in research grade electrodes (e.g. BrainProducts, Neuroscan).

The founder's previous research seems to be focused on topics related to consciousness and sleep. I was hoping if he published articles about electrode materials, but can't find any. Maybe it's a trade secret.. Anyone with more info on this?

Also, their electrodes seem to be mostly focused on the occipital area. I get the impression they'd be focused more on visual components rather than cognition and decision making (e.g. P300).

Funny you'd mention that, there was a study from Jonas and Kording [0] that considered a microprocessor as an organism and applied analytic methods used in neuroscience to see if they can figure out how it processes information.

[0] Jonas, E. and Kording, K.P., 2017. Could a neuroscientist understand a microprocessor?. PLoS computational biology, 13(1), p.e1005268.

https://doi.org/10.1371/journal.pcbi.1005268